If `FlappingRestartIntervalHours` for OSD is > 0 in the cephCluster CR,
then start OSD inside a bash script.
Signed-off-by: sp98 <sapillai@redhat.com>
(cherry picked from commit ffd69291f1)
Using the new configuration parameters prometheusEndpoint and
prometheusEndpointSSLVerify users can configure dashboard to
point to their prometheus instance.
closes: #12876
Signed-off-by: Redouane Kachach <rkachach@redhat.com>
(cherry picked from commit c763935436)
When OSDs flap, ceph stops the OSD daemon if its marked down greater than
5 times in 600 seconds. But OSD pod restarts and marks the OSD `up` again.
This causes the PGs mapped to these OSDs to peer. While the PGs are peering,
IO to these PGs are blocked.
So we need to ensure that if ceph is marking OSD `down` due to flapping, OSD pod
should not restart to mark the OSDs `up` again.
This PR adds a sleep to the OSD pod if the container returned with a 0 exit code
Default behavior is to sleep for 6 hrs. But user can configure it from the
ceph cluster spec.
Signed-off-by: sp98 <sapillai@redhat.com>
(cherry picked from commit 4eb9f62205)
While two mgrs is considered sufficient, more mgr daemons are
now allowed in case the admin wants even more fault tolerance
to the active mgr going down, in case multiple mgrs go down.
Up to five mgr pods will be allowed. All mgrs will be in
standby mode except one active mgr. All mgr pods have a sidecar
that will update their respective pod specs with the active
or passive label.
Signed-off-by: travisn <tnielsen@redhat.com>
(cherry picked from commit 66547d27be)
Change how Rook detects network CIDRs for Multus networks. The IPAM
configuration is only defined as an arbitrary string JSON blob with a
"type" field and nothing more. Rook's detection of CIDRs for whereabouts
had already grown out of date since the initial implementation.
Additionally, Rook did not support DHCP IPAM, which is a reasonable
choice for users. And more, Rook did not support CNI plugin chaining,
which further complicates NADs. Based on the CNI spec, network chaning
can result in any changes to network CIDRs from the first-given plugin.
All these problems make it more and more difficult for Rook to support
Multus by inspecting the NAD itself to predict network CIDRs. Instead,
it is better for Rook to treat the CNI process as a black box. To
preserve legacy functionality of auto-detecting networks and to make
that as robust as possible, change to a canary-style architecture like
that used for Ceph mons, from which Rook will detect the network CIDRs
if possible.
Also allow users to specify overrides for CIDR ranges. This allows Rook
to still support esoteric and unexpected NAD or network configurations
where a CIDR range is not detectable or where the range detected would
be incomplete. Because it may be impossible for Rook to understand the
network CIDRs wholistically while residing only on a portion of the
network, this feature should have been present from Multus's inception.
Improving CIDR auto-detection and allowing users to specify overrides
for auto-detected CIDRs rounds out Rook's Multus support for CephCluster
(core/RADOS) installations. No further architectural changes should be
needed for CephClusters as regards application of public/cluster network
CIDRs for Multus networks.
Signed-off-by: Blaine Gardner <blaine.gardner@ibm.com>
(cherry picked from commit 3c43268d0a)
The object user was previously required to be created in the
same namespace as the object store and the cluster. Now,
the object user can be reconciled even in a different namespace
from the cluster and object store. The namespace would be specified
in the object user CR.
Signed-off-by: travisn <tnielsen@redhat.com>
(cherry picked from commit 2b1cc4c7b8)
The go package dependencies (go.mod and go.sum) were created
under the pkg/apis directory to simplify the dependencies for
other projects referencing the rook repo. The downside is that
the dependabot can no longer open a working and valid PR since
the bot is not aware of the go.mod and go.sum in the apis directory.
Since the reduction of dependencies for vault in #12455,
the list of extra dependencies is not quite so extensive.
We are putting effort into shrinking the dependency list instead
of using the modules files in the apis subdirectory.
Note that #12419 updated the dependencies for the latest controller
runtime which also increased the size of the modules
list in the pkgs subdirectory, which means the difference
in dependencies in the pkgs subdirectory is no longer
signficant anyway.
Signed-off-by: travisn <tnielsen@redhat.com>
(cherry picked from commit 1a4917de78)
# Conflicts:
# .github/workflows/update_dependabot_bundler_pr.yml
If osd store is updated in the ceph cluster, then
delete OSDs one by one, cleanup disks and provision a new OSD on
the same disk
Signed-off-by: sp98 <sapillai@redhat.com>
(cherry picked from commit 886bb357ef)
golangci linter was throughing error `k8s.io/utils/pointer`
package is deprecated. So, I have removed that.
Signed-off-by: subhamkrai <srai@redhat.com>
(cherry picked from commit a25071ac29)
update vault and other indirect imports. Also,
updating k8s package to `v0.28.0-rc.0` since with
v0.27.4 it was through build error.
Signed-off-by: subhamkrai <srai@redhat.com>
(cherry picked from commit ab3763767f)
users & buckets admin capabilities aren't spelled the same
way between rook crd (singular) and ceph (plural). It's a bit
misleading when comparing ceph admin cap and rook users.
Signed-off-by: Peter Goron <peter.goron@gmail.com>
(cherry picked from commit 40eed236cf)
Adding CephCOSIDriver CRD and controller. The controller will bring up
the ceph cosi driver when first object store is created in the rook
operator namespace. Then admin can defined COSI CRDs like BucketClass
and BucketAccessClass for different object stores deployed via Rook.
Using the BucketClass and BucketAccessClass, user can define
BucketAccess for backend bucket in the RGW. The CephCOSIDriver CRD
defines configuration options for ceph cosi driver. In the first version
its usability is minimal. Even if it is not defined Rook will bring up
the ceph cosi driver with default values.
Signed-off-by: Jiffin Tony Thottan <thottanjiffin@gmail.com>
(cherry picked from commit b48dc8a335)
Some users want to deploy two CephObjectStores for a single Zone. The first
configures RGWs to process the synchronization of the data, while the second
CephObjectStore configures the client RGWs.
Currently, this can be implemented by setting the RGW option
'rgw_run_sync_thread' in the 'rook-config-override' ConfigMap, though it is not
really user friendly.
Ref: https://docs.ceph.com/en/latest/radosgw/config-ref/#confval-rgw_run_sync_thread
This commit adds a new option in the CephObjectStore CRD as defined in issue
https://github.com/rook/rook/issues/12272. The new option
'disableMultisiteSyncTraffic' determine whether the operator should disable the
multisite sync threads for the RGWs.
If the option is set to 'false', or if the option is not specified, the operator
does nothing. This ensures that the multisite sync threads will not be enabled
for users that disabled explicitely the multisite sync threads either manually
or with the 'rook-config-override' ConfigMap.
This commit also recommends to use two objectstore when scaling Ceph Objectstore
Multisite replication, with one objectstore configured with disabled replication
traffic.
Signed-off-by: Lucas Henry <polyedre@disroot.org>
(cherry picked from commit 393d09347e)
in the existing node watcher, we'll check for node update
event and see if there are `out-of-service` taints are applied
and `ROOK_WATCH_FOR_NODE_FAILURE` is enabled in rook-ceph-operator-configmap,
if then we'll create the networkFence cr and delete the cr if nodes come back.
And, added the unit test too.
Signed-off-by: subhamkrai <srai@redhat.com>
The vault server dependencies were pulling a lot of indirect modules
that are not necessary for Rook. The modules were only used
for testing. The vault github explicitly indicates that:
"Some other projects have made a practice of doing so in order to
take advantage of testing tooling that was developed for testing Vault itself.
This is not, and has never been, a supported way to use the Vault project."
When attempting to update to vault v1.13.3, we are hitting
such an unsupported configuration with go modules
that are impossible to satisfy. By removing the tests
that were using the vault server project, we can remove
this large set of dependencies, enable updating to v1.13.3,
and also reduce the modules pulled in by vault.
Signed-off-by: travisn <tnielsen@redhat.com>
with `requiredDropCapabilities: ["All"]` we can
drop all the default privileges and we can add the
privileges required `allowedCapabilities:` here.
Signed-off-by: subhamkrai <srai@redhat.com>
The go modules are updated to the latest client-go, vault,
and other modules. There does not appear to be a combination
of modules that will allow us to use vault v1.13.3, so we
stick with v1.13.2.
Signed-off-by: travisn <tnielsen@redhat.com>
We add a new field domainName to the Kerberos section. The field is used
to setup /etc/idmapd.conf with the domain name. This allows idmapper to
map to kerberos credential to the correct uid/gid.
We add Spec.Security.Kerberos.DomainName to the CRD
Signed-off-by: Sachin Prabhu <sprabhu@redhat.com>
1) donot change rgw fqdn to ip if provided,
As now the bucket class supports the
entry of fqdn
2) update crds with new description in EndpointAddress
Signed-off-by: parth-gr <paarora@redhat.com>
The prometheus mgr module and ceph exporter can now be optionally
disabled by the monitoring.metricsDisabled setting in the
CephCluster CR. These will not be disabled by default, rather
than the mgr module being disabled by default from v1.11.4.
Signed-off-by: travisn <tnielsen@redhat.com>
Pare down the available volume sources for NFS config files so that the
CRD isn't unnecessarily huge. This allows us to recommend
`kubectl apply` again in the upgrade doc.
Size of the NFS CRD is reduced by approximately 60%.
Signed-off-by: Blaine Gardner <blaine.gardner@redhat.com>
The journal size was only applicable to the filestore OSD
format which has not been supported by rook since v1.2.
Remove the remaining obsolete setting from the examples and
code.
Signed-off-by: travisn <tnielsen@redhat.com>
ClusterID uniquely identifies a cluster. It is used as a prefix to
nslookup exported services.
For example: <clusterid>.<svc>.<ns>.svc.clusterset.local
Signed-off-by: sp98 <sapillai@redhat.com>
The defaults in CRDs cause an update and trigger a new
reconcile if the value is not set. For the key rotation
setting, the default was only being updated when the kms
feature was enabled. This update was causing the reconcile
to fail in the kms tests. With the default now in code,
the default can remain empty and no cr update will trigger
a new reconcile unexpectedly.
Signed-off-by: travisn <tnielsen@redhat.com>
The rgw dashboard can be disabled by the setting in the
CephObjectStore `gateway.dashboardEnabled` if set to false.
Signed-off-by: xiaobaowen <xiaobaowen@deeproute.ai>
validate the object store name only if it is not an external cluster
as external cluster won't create the rgw daemon and it's other resource
and there is some legacy external cluster which has more lenght of objectstore
so to run them successfully we are not validating the objectstore name
Signed-off-by: parth-gr <paarora@redhat.com>
This commit adds util function to create key rotation cron job
such as getKeyRotationContainer, getKeyRotationCronPodTemplateSpec
and makeKeyRotationCronJob.
Signed-off-by: Rakshith R <rar@redhat.com>
This commit adds KeyRotationSpec to cephcluster CR
to support encryption key rotation feature.
If enabled Rook will create a cronjob for each
encrypted PVC based OSD with given schedule set
to rotate their respective key encryption key.
Signed-off-by: Rakshith R <rar@redhat.com>
Add a hostNetwork option for a Ganesha server.
This allows you to use host networking for the cluster but still
use regular pod networking for NFS Ganesha servers
Signed-off-by: Richard Bateman <taxilian@gmail.com>